Introduction Questions and Themes The idea of equality is confronted by two different types of diversities." (1) the basic heterogeneity of human beings, and (2) the multiplicity of variables in terms of which equality can be judged. This book is concerned with both these diversities. It is also specifically concerned with the relation between the two. The heterogeneity of people leads to divergences in the assessment of equality in terms of different variables. This adds significance to the central question: equality of what? Diverse Humanity Human beings are thoroughly diverse. We differ from each other not only in external characteristics (e.g. in inherited fortunes, in the natural and social environment in which we live), but also in our personal characteristics (e.g. age, sex, proneness to illness, physical and mental abilities). The assessment of the claims of equality has to come to terms with the existence of pervasive human diversity, The powerful rhetoric of 'equality of man* often tends to deflect attention from these differences. Even though such rhetoric (e.g. "all men are born equal*) is typically taken to be part and parcel of egalitarianism, the effect of ignoring the interpersonal variations can, in fact, be deeply inegalitarian, in hiding the fact that equal consideration for all may demand very unequal treatment in favour of the disadvantaged. The demands of substantive equality can be particularly exacting and complex when there is a good deal of antecedent inequality to counter. Sometimes, human diversities are left out of account not on the misconceived 'high* ground of 'equality of human beings", but on the pragmatic low* ground of the need for simplification. But the net result of this can also be to ignore centrally important features of demands of equality.
2 Introduction: Questions and Themes Diversity of Foots Equality is judged by comparing some particular aspect of a person (such as income, or wealth, or happiness, or liberty, or opportunities, or rights, or need-fulfilments) with the same aspect of another person. Thus, the judgement and measurement of inequality is thoroughly dependent on the choice of the variable (income, wealth, happiness, etc.) in terms of which comparisons are made, I shall call it the 'focal variable' the variable on which the analysis focuses, in comparing different people, The chosen focal variable can, of course, have an internal plurality. For example, freedoms of different types may be put together as the preferred focus of attention, or the variable selected may involve a combination of freedoms and achievements. The multiple features within a chosen focal variable have to be distinguished from the diversity between the chosen focal variables. Some variables that are often taken to be elementary and uniform do, in fact, have much internal plurality (e.g. real income or happiness).' To use the kind of language for which we economists are often not unreasonably teased, this is the question of the choice of 'space' in which different persons are to be compared. That spatial analogy, despite its demonstratively Cartesian pretensions, is a useful classiflcatory device, and I shall invoke it to separate out the problem of the choice of focal variables ('the choice of space') from other issues in the assessment of inequality, Links and Disharmonies The characteristics of inequality in different spaces (such as income, wealth, happiness, etc,) tend to diverge from each other, because of the heterogeneity of people. Equality in terms of one variable may not coincide with equality in the scale of another. For example, equal opportunities can lead to very unequal incomes. Equal incomes can go with significant differences in wealth. Equal wealth can coexist with very unequal happiness. Equal happiness can go with widely divergent fulfilment of needs. Equal fulfilment of needs can be associated with very different freedoms of choice. And so on. If every person were much the same as every other, a major cause 1 I have discussed these issues elsewhere, addressing also the problem of overall ranking and aggregate valuation of inherently plural variables (in Sen 1980-1,!982a).
Introduction: Questions and Themes 3 of these disharmonies would disappear. If the rankinp of equality in different spaces coincide, it would then be less important to have a clear answer to the question: equality of what? The pervasive diversity of human beings intensifies the need to address the diversity of focus in the assessment of equality, Diverse EgaUtarianism It is convenient to begin with the observation that the major ethical theories of social arrangement all share an endorsement of equality in terms of some focal variable, even though the variables that are selected are frequently very different between one theory and another. It can be shown that even those theories that are widely taken to be 'against equality* (and are often described as such by the authors themselves) turn out to be egalitarian in terms of some other focus. The rejection of equality in such a theory in terms of some focal variable goes hand in hand with the endorsement of equality in terms of another focus. For example, a libertarian approach (such as the entitlement theory forcefully developed in Robert Noziek's Anarchy, State and Utopia 2 ) may give priority to extensive liberties to be equally guaranteed to each, and this demands rejecting equality or any 'patterning* of end states (e.g. the distribution of incomes or happiness). What is taken usually by implication to be a more central focus rules the roost, and inequalities in the variables that are, in effect, treated as peripheral must, then, be accepted in order not to violate the right arrangements (including equality) at the more central level. Plausibility and Equality There is a reason for this apparently ubiquitous 'egalitarianism'. Ethical plausibility is hard to achieve unless everyone is given equal consideration in some space that is important in the particular theory (Chapter 1). While it may be too ambitious to claim (as some have done) that this is a logical necessity, or simply a part of the discipline of the language of morals, 3 it is difficult to see how an 2 See Noaek (1973, 1974). For a reassessment and some revision, see Nozick (1989). 3 See particularly Hare (1952, 1963).
4 Introduction: Questions and Themes ethical theory can have general social plausibility without extending equal consideration to all at some level. While the question 'why equality?' is by no means dismissible, it is not the central issue that differentiates the standard theories, since they are all egalitarian in terms of some focal variable. The engaging question turns out to be 'equality of what?' To that question 'equality of what?* -different theories give different answers. The different answers are distinguishable in principle and involve different conceptual approaches. But the practical force of these distinctions depends on the empirical importance of the relevant human heterogeneities which make equality in one space diverge from equality in another. Achievement and Freedom Sources of divergence between different approaches can, of course, go well beyond the identification of the space itself, and may be concerned instead with the way the space is utilized. In the standard theory of inequality measurement, these problems of "appropriate indices* have tended to receive much attention. The analysis can fruitfully proceed on the basis of postulating explicitly or implicitly acceptable axioms for inequality assessment in that space. While the focus of this book is on the choice of space and its implications, it is not my intention to deny the practical importance of these indexing problems in a given space (this was in fact the main subject of analysis in my previous book on inequality 4 ). One of the aspects of inequality assessment that has received less attention than it deserves relates to the distinction between achievement and the freedom to achieve. The nature, reach, and relevance of that distinction between achievement and freedom are briefly discussed in Chapter 2, making use of conceptual grounds for discrimination as well as analytical procedures used in modern economics. Functionings and Capability The monograph then proceeds to identify, develop, and defend a particular choice of space and its use in terms of the freedom to achieve (Chapter 3). A person's capability to achieve functionings 4 On Economic Inequality; Sen 1973«in bibliography. Since I shall have to cite it frequently (mainly to avoid having to repeat myself), I shall refer to it in a more easily recognizable form, viz, OBI.
Introduction; Questions and Themes 5 that he or she has reason to value provides a general approach to the evaluation of social arrangements, and this yields a particular way of viewing the assessment of equality aad inequality, The functionings included can vary from most elementary ones, such as being well-nourished, avoiding escapable morbidity and premature mortality, etc,, to quite complex and sophisticated achievements, such as having self-respect, being able to take part in the life of the community, and so on. The selection and weighting of different functionings influence the assessment of the capability to achieve various alternative functioning bundles. The roots of this approach can be traced to Aristotelian distinctions, but its ramifications can take various different forms. The particular class of possibilities developed here is less assertive and less insistently complete than some possible alternatives. But it is also less demanding on interpersonal agreement and more tolerant of unresolved disputes. Evaluation of Effective Freedom The concentration on the freedom to achieve and not just on the level of achievement raises some deep questions about the connection between the appraisal of the alternative achievements and the value of the freedom to achieve them (Chapter 4). Even the freedom-based perspective must pay particular attention to the nature and value of the actual achievements, and inequalities in achievement can throw light on inequalities in the respective freedoms enjoyed. This recognition requires us to reject such proposed rules of freedom assessment as the counting of the number of alternatives in the 'range of choice'. More constructively, it suggests practical ways of using observable data regarding achievements to get a partial but significant view of the freedoms enjoyed by different persons. In this context I also discuss the difference between well-being objectives and the other objectives a person may have. This difference not only leads to some plurality within the idea of freedom itself, it also has important implications on the divergence between the perspective of achievements and that of freedoms. One of the related issues taken up here is the possibility that more freedom can be disadvantageous to" a person, which if generally true can undermine the rationale of judging inequality in terms of freedoms. I argue that the real conflict is between different types of
6 Introduction: Questions and Themes freedoms, and not between freedom tout court and advantages in general. Distinctions: Capability and Utility The focus on the space of functionings and on the capability to achieve functionings differs quite substantially from the more traditional approaches to equality, involving concentration on such variables as income, wealth, or happiness (Chapters 3 and 4). The fact of human diversity is closely related to substantive conflicts between focusing on different informational bases for assessing equality, efficiency, and justice. In particular, judging equality and efficiency in terms of the capability to achieve differs from the standard utilitarian approaches as well as from other welfarist formulations. Welfarism in general and utilitarianism in particular see value, ultimately, only in individual utility, which is defined in terms of some mental characteristic, such as pleasure, happiness, or desire, 5 This is a restrictive approach to taking note of individual advantage in two distinct ways; (I) it ignores freedom and concentrates only on achievements, and (2) it ignores achievements other than those reflected in one of these mental metrics. In so far as utility is meant to stand for individual well-being, it provides a rather limited accounting of that, and it also pays no direct attention to the freedom to pursue well-being or any other objective (Chapter 3). This way of seeing individual advantage is particularly limiting in the presence of entrenched inequalities. In situations of persistent adversity and deprivation, the victims do not go on grieving and grumbling all the time, and may even lack the motivation to desire a radical change of circumstances. Indeed, in terms of a strategy for living, it may make a lot of sense to come to terms with an ineradic- 5 There is some ambiguity in the characterization of the 'preference* view of utility, as it can be and has been defined in quite distinct and divergent ways. If it is defined entirely in terms of individual choice (as in Samuelson 1938), then the preference view does not give any immediate content to interpersonal comparisons and thus does not yield any straightforward judgements of inequality. The devised meanings that can be somewhat artificially constructed (e.g. Harsanyi's 1955 engaging proposal that we consider our preference regarding who we would choose to be) involve conceptual problems as well as empirical difficulties (I have discussed this issue in Sen 1982a), On the other hand, if preference is defined in terms of a person's sense of desire or satisfaction (as in Edgeworth 1881 or Hicks 1939), then the preference approach is in line with the mental-metric views considered in the text.
INEQUALITY REEXAMINED AMARTYA SEN RUSSELL SAGE FOUNDATION NEW YORK CLARENDON PRESS OXFORD
This book has been printed digitally a"d produced in a standard specification in order to ensure its continuing availability OXPORD UNIVERSITY PRESS Great Clarendon Street, Oxford OX2 6DP Oxford University Press is a department of the University of Oxford. It furthers the University's objective of excellence in research, scholarship, and education by publishing worldwide in Oxford New York Auckland Cape Town Dar es Salaam Hong Kong Karachi Kuala Lumpur Madrid Melbourne Mexico City Nairobi New Delhi Shanghai Taipei Toronto With offices in Argentina Austria Brazil Chile Czech Republic France Greece Guatemala Hungary Italy Japan South Korea Poland Portugal Singapore Switzerland Thailand Turkey Ukraine Vietnam Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries Published in the United States by Oxford University Press Inc., New York Amartya Sen 1992 The moral rights of the author have been asserted Database right Oxford University Press (maker) Reprinted 2006 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, op>r as expressly permitted by law, or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above You must not circulate this book in any other binding or cover And you must impose this same condition on any acquirer fsbn 0-19-828334-2